Modeling and real-time prediction of irrigation water quality index using Graphical User Interface (App), a case study of Oued-Hammam North-East Algeria

Abstract

Abstract The quality of water for irrigation purposes hasnot been taken into serious consideration in many places around the Globe. The present case study at Oued-Hammam watershed aims to investigate the use of an artificial neural network in the prediction of irrigation water quality indicators of Sodium absorption ratio and Electrical Conductivity and deployment of the models using graphical user interface (App).Fourteen water quality parameters were collected at Zit Emba reservoir from 2010 to 2014.Pearson correlation matrix was used to select input parameters with respect with the output parameter. The back-propagation neural networks learning algorithm was used in modeling of irrigation water quality index (IWQI) for both SAR and EC. The performances of models were evaluated using statistical criteria of correlation coefficient (R) and root mean square error (RMSE). Back propagation neural network learning algorithm maximum correlation coefficient for SAR and EC were 0.98077 and 0.97762 respectively, also with minimum RMSE of 0.037 for SAR and 101.8 for EC. Thus current study suggests that artificial neural network (ANN) models are most effective tools for prediction of water quality prediction and their outcome can be used as effective method in management and real-time control of water pollution around the watershed.